Pre-clinical safety evaluation for mRNA-vaccine development in mouse model

Messenger RNA (mRNA) vaccine has emerged as an attractive agent for prevention of infectious disease and anti-cancer therapy. However, there is a fatal risk that the safety evaluation for mRNA vaccine have not been fully studied yet. In this study, we evaluated the safety of four type of COVID-19 S-protein targeting mRNA vaccines with different compositions (C2/ LNP90, C2LNP128, C3LNP90 and C3LNP128). Theses vaccines were intramuscularly injected to 6-wk old male and female ICR mice with twice at an interval of 2 wks. The necropsy was carried out on 2 days or 14 days after secondary injection. The results showed that the body weight was decreased for 2days after the first injection in C2/LNP128 and C3/LNP128-injected mice, but it was almost recovered at 7day post injection (dpi). At 2 dpi after secondary injection, the endpoint blood analysis of demonstrated that C2/LNP128 and C3/LNP128 decreased the number of lymphocytes, monocytes and reticulocytes carrying the abnormal level of liver function indicator such as albumin, AST, ALT and total protein. Additionally, C2/LNP128 decreased the number of platelet and C3LNP128 decreased the number of red blood cells, respectably. Spleen and inguinal lymph node were enlarged in all experimental group. Notably, C2/LNP128 and C3/LNP128 induced severe edema in injection site, femoris muscle. At 14 dpi after secondary injection, the toxicity that was observed at 2 dpi after secondary injection was recovered. These results suggest that the potential side effects of mRNA vaccines must be systematically evaluated with multiple aspect of toxicology.

Mixing-AdaSIN: Constructing a De-biased Dataset Using Adaptive Structural Instance Normalization and Texture Mixing

Following the pandemic outbreak, several works have proposed to diagnose COVID-19 with deep learning in computed tomography (CT);reporting performance on-par with experts. However, models trained/tested on the same in-distribution data may rely on the inherent data biases for successful prediction, failing to generalize on out-of-distribution samples or CT with different scanning protocols. Early attempts have partly addressed bias-mitigation and generalization through augmentation or re-sampling, but are still limited by collection costs and the difficulty of quantifying bias in medical images. In this work, we propose Mixing-AdaSIN;a bias mitigation method that uses a generative model to generate de-biased images by mixing texture information between different labeled CT scans with semantically similar features. Here, we use Adaptive Structural Instance Normalization (AdaSIN) to enhance de-biasing generation quality and guarantee structural consistency. Following, a classifier trained with the generated images learns to correctly predict the label without bias and generalizes better. To demonstrate the efficacy of our method, we construct a biased COVID-19 vs. bacterial pneumonia dataset based on CT protocols and compare with existing state-of-the-art de-biasing methods. Our experiments show that classifiers trained with de-biased generated images report improved in-distribution performance and generalization on an external COVID-19 dataset. © 2021, Springer Nature Switzerland AG.

Status of Ambient PM2.5 Pollution in the Seoul Megacity (2020)

The Center for Air Quality &Control at the Seoul Research Institute of Public Health and the Environment (SIHE) has monitored changes in the concentration of fine dust in Seoul over the past 10 years and investigated meteorological factors as well as fine particulate matter (PM2.5), sulfur dioxide (SO2), and nitrogen dioxide (NO2) concentrations in northeastern China and its contribution to the PM2.5 concentration in Seoul. The concentration of fine dust in Seoul in 2020 was 21 µg/m3, which is down 16% from 2019 and the lowest since 2010. In 2020, China’s emissions of pollutants such as NO2 have decreased significantly due to regional blockades, social distancing, and factory shutdowns caused by COVID-19. As a results, the concentration of precursors such as SO2 and NO2, and PM2.5 in northeastern China are also decreased, which contributed to the reduction in PM2.5 concentration in Seoul caused by westerly winds blowing. In addition, the ratio of east and south winds that usually contain low concentrations of pollutants was more than 30% of the total air currents into Seoul, which is the highest in the last three years. Moreover, the mean wind velocity and the amount of precipitation were also the highest recorded values of 2.4 m/s and 1651.0 mm, respectively. Calculations using Comprehensive Air quality Model with eXtensions (CAMx)-Particulate Source Apportionment Technology (PSAT) show that the contribution of external inflows to the PM2.5 concentration in Seoul was 65%. We believe that the reasons for the low PM2.5 concentration in 2020 are due to meteorological factors and a decrease in air pollution in northeastern China. Meanwhile, the major contribution of emissions in Seoul (resuspended road dust and non-exhaust dust) was high. When the concentration of PM2.5 was high, the contribution of resuspended road dust was reduced due to an increase of secondary generating materials. Currently, data on emission reduction due to the COVID-19 cannot be assessed, which we believe will enable more accurate contribution calculations in the future. © 2021